MicroRNA-29b induces global DNA hypomethylation and tumor suppressor gene reexpression in acute myeloid leukemia by targeting directly DNMT3A and 3B and indirectly DNMT1

Abstract

Aberrant DNA hypermethylation contributes to myeloid leukemogenesis by silencing structurally normal genes involved in hematopoiesis. MicroRNAs (miRNAs) are noncoding RNAs that regulate gene expression by targeting protein-coding mRNAs. Recently, miRNAs have been shown to play a role as both targets and effectors in gene hypermethylation and silencing in malignant cells. In the current study, we showed that enforced expression of miR-29b in acute myeloid leukemia cells resulted in marked reduction of the expression of DNA methyltransferases DNMT1, DNMT3A, and DNMT3B at both RNA and protein levels. This in turn led to decrease in global DNA methylation and reexpression of p15(INK4b) and ESR1 via promoter DNA hypomethylation. Although down-regulation of DNMT3A and DNMT3B was the result of a direct interaction of miR-29b with the 3' untranslated regions of these genes, no predicted miR-29b interaction sites were found in the DNMT1 3' untranslated regions. Further experiments revealed that miR-29b down-regulates DNMT1 indirectly by targeting Sp1, a transactivator of the DNMT1 gene. Altogether, these data provide novel functional links between miRNAs and aberrant DNA hypermethylation in acute myeloid leukemia and suggest a potentially therapeutic use of synthetic miR-29b oligonucleotides as effective hypomethylating compounds.

Figures

Figure 1

miR-29b targets DNMT1, DNMT3A, and 3B in AML cell lines and primary AML blasts. (A) DNMT3A, (B) DNMT3B, and (C) DNMT1 mRNA expression after nucleoporation (K562, MV4-11) or infection (Kasumi-1) of pre-miR-29b or their respective controls (scrambled oligonucleotides for K562 and MV4-11cells and empty vector lentivirus for Kasumi-1 cells). Histograms show fold changes (reduction) in mRNA expression with respect to the controls after normalization with 18s in 3 independent experiments. Bars represent SD. (D) DNMT1, DNMT3A, and 3B protein expression after pre-miR-29b or control nucleoporation or infection in K562, MV4-11, and Kasumi-1 cell lines. Equivalent gel loading was confirmed by probing with antibodies against β-actin. (E) DNMT1 and DNMT3A and 3B mRNA expression in primary AML patient samples (n = 3) after nucleoporation of pre-miR-29b or a scrambled oligonucleotide. Histograms show fold changes (reduction) in mRNA expression with respect to the control after normalization with 18s. Bars represent SD.

Figure 2

DNMT3A and 3B are targets of miR-29b. (A) Schema of the 2 firefly luciferase reporter constructs for DNMT3A and 3B, indicating the interaction sites between miR-29b and the 3′ UTRs of the DNMT3s. (B) Dual luciferase assay of K562 cells cotransfected with firefly luciferase constructs containing the DNMT3A or 3B wild-type or mutated 3′ UTRs and pre-miR-29b or scrambled oligonucleotides as indicated. The firefly luciferase activity was normalized to Renilla luciferase activity. The data are shown as relative luciferase activity of pre-miR-29b transfected cells with respect to the control (scrambled oligonucleotide) of a total of 9 experiments from 3 independent transfections. Bars represent SD. (C) Dual luciferase assay of K562 cells cotransfected with firefly luciferase constructs containing the DNMT1 3′ UTR and pre-miR-29b or scrambled oligonucleotides. The data are shown as relative luciferase activity of pre-miR-29b–transfected cells with respect to the control (scrambled oligonucleotide) of 9 experiments obtained from 3 independent transfections. Bars represent SD.

Figure 3

miR-29b targets Sp1. (A) Genomic representation of the Sp1 3′ UTR with the localization of the 4 predicted pre-miR-29b binding sites. Immediately below there is a schema of the luciferase reporter assays used in the luciferase experiments. The “X” sign over the red boxes indicates the miR-29b sites that were deleted. mut indicates mutated. (B) Dual luciferase assay of K562 cells cotransfected with firefly luciferase constructs containing the wild-type or mutants target site of the Sp1 3′ UTR region with pre-miR-29b or a scrambled oligonucleotide. The data are shown as relative luciferase activity of pre-miR-29b-transfected cells with respect to the control (scrambled oligonucleotide) of 9 experiments from 3 independent transfections. Bars represent SD. (C) Sp1 mRNA and protein expression after nucleoporation (K562, MV4-11) or infection (Kasumi-1) of pre-miR-29b or their respective controls, scrambled oligonucleotide, or empty vector lentivirus. Histograms show fold changes with SD. (D) Sp1 mRNA expression in 3 primary AML samples after nucleoporation of pre-miR-29b or scrambled oligonucleotide. Histograms show fold change (reduction) with respect to control after normalization with 18s. Bars represent SD.

Figure 4

DNMT1 promoter activity was down-regulated by pre-miR-29b through repression of Sp1. (A) EMSA of DNMT1 promoter DNA and nuclear extracts prepared from K562 cells transfected with miR-29b construct (miR-29b) or scrambled control (sc). The arrow represents the DNMT1-protein complex. (B) EMSA of DNMT1 promoter DNA with human recombinant Sp1 protein (0.5 mg, rhSp1, Promega, catalog no. E6391) or no protein extracts. These data further indicate that Sp1 is binding to the DNMT1 promoter DNA. (C) EMSA of DNMT1 promoter DNA using K562 nuclear extracts with unlabeled DNA competitors (Sp1-binding elements and DNMT1 promoter DNA, WT). As a control, we used no DNA competitor (−) and a nonspecific unlabeled probe (TFIID) that contains the TATA box sequence 5′-GCAGAGCATATAAGGTGAGGTAGG A-3′. This sequence is not related to the DNMT1 promoter DNA. The fact that this oligo did not compete shows specificity of DNA competition by DNMT1 DNA and by SP1 DNA. (D) EMSA of DNMT1 promoter DNA using K562 nuclear extracts and unlabeled wild-type (WT) or mutants (M) excess (20 times) DNA competitors. The first 2 lanes (from the left) are without competitors. The sequences of the DNMT1 promoter DNA and various DNMT1 mutants with linker scanned mutations (shown in bold-type) are shown below (M1-M5).

Figure 5

Overexpression of pre-miR-29b in AML cell lines reduces GDM. MV4-11 (A) and Kasumi-1 (B) cell lines were nucleoporated (MV4-11) or infected (Kasumi-1) with pre-miR-29b (oligonucleotides and lentivirus) or their respective controls. DNA was obtained from both cell lines after 72 hours of nucleoporation or infection, and GDM was measured by LC-MS/MS. Results from treatment with 2.5 μM decitabine, a hypomethylating agent, or phosphate-buffered saline (control) are also shown for the MV4-11 cell line as positive controls. The data are shown as absolute GDM. Bars represent range of 2 independent experiments.

Figure 6

Pre-miR-29b restores expression of hypermethylated ESR1 and p15INK4b in the MV4-11 cell line. (A) Quan-titative RT-PCR for ESR1 in K562 and MV4-11 cells (B), and p15INK4b (C) after 72 hours of nucleoporation with pre-miR-29b or scrambled oligonucleotides. Histograms represent mean values with SD. (D) Scatter plots of the quantitative DNA methylation data of ESR1 promoter regions in K562 (D) and MV4-11 cells (E) after nucleoporation with pre-miR-29b (■) or a scrambled oligonucleotide (●) obtained using the MassArray system. The lines connecting the dots indicate the corresponding CpG pairs. (F) Scatter plots of the quantitative DNA methylation data of p15INK4b in MV4-11 cells after nucleoporation with pre-miR-29b (■) or a scrambled oligonucleotide (●). Each dot represents a single CpG or a group of CpGs analyzed. The lines connecting the dots indicate the corresponding CpG pairs. The scatter plots were created using GraphPad Prism, version 5.00 for Windows (GraphPad Software, San Diego, CA). The P values were obtained comparing the CpG methylation levels between scrambled and pre–miR-29b–transfected cells using paired Wilcoxon signed-rank test. Error bars indicate SD.

Figure 7

Overexpression of pre-miR-29b induces partial differentiation of AML blasts. (A) Flow cytometry analysis of CD11b expression in Kasumi-1 cells infected with lentivirus-expressing miR-29b or empty vector. Green represents empty vector (EV); red, miR-29b lentivirus.